Previous studies in our laboratory have demonstrated that both CD8+ and CD4+ T cells, derived from infiltrates in melanomas and others types of tumors, can manifest MHC-restricted recognition of tumor-associated antigens (Ag). MHC class I-restricted Ag recognized by CD8+ cytolytic T cells have been targeted by experimental vaccines for melanoma and other types of cancers, with limited success. Current projects in our laboratory focus on identifying melanoma associated proteins recognized by CD4+ helper T cells for developing more effective melanoma vaccines, as well as on extending this work to prostate cancer. 1) SEARCH FOR TUMOR-ASSOCIATED PROTEINS RECOGNIZED BY CD4+ T CELLS. In efforts to develop efficient and sensitive methods for identifying MHC class II-resticted tumor Ag recognized by helper T cells, a number of strategies have been pursued. Previously, a biochemical purification approach was developed to isolate a mutated glycolytic enzyme, triosephosphate isomerase, as the unique tumor Ag recognized by CD4+ T cells from a melanoma patient. The effects of this mutation on enzyme function and possibly tumorigenesis are currently being investigated. More recently, in collaboration with other investigators in the Surgery Branch, a molecular cloning strategy has been developed that depends on targeting proteins expressed by a tumor-derived cDNA library to the endosomal processing compartment,using invariant chain fusion. The first tumor Ag to be identified with this system was a mutated CDC27, derived from a melanoma. We are currently characterizing another Ag identified with this system, which appears to be a novel tumor-associated Ag. 2) CLINICAL EVALUATION OF TYROSINASE AS AN IMMUNOGEN AGAINST MALIGNANT MELANOMA. Our laboratory has identified the melanoma associated protein, tyrosinase, as a tumor Ag recognized by both cytolytic (CD8+) T cells and helper (CD4+) T cells. As such, it may prove to be a potent immunogen against melanoma. This is being evaluated in clinical protocol #99-C-0095, "Immunization of patients with metastatic melanoma using recombinant vaccinia and fowlpox viruses encoding the tyrosinase antigen", which is one of the first trials to use two different poxvirus vectors in a heterologous prime/boost format. Although clinical response is the primary endpoint of this trial, sera and lymphocytes are being collected from patients at intervals to assess IgG and T cell responses against tyrosinase which may be generated through vaccination. To date, 31 patients have been randomized on this trial. Two patients among 18 tested have generated serum IgG responses against tyrosinase following vaccination. 3) DEFINING IMMUNE RESPONSES AGAINST PROSTATE CANCER. Limited information is available on the human immune response to prostate cancer, in part due to a scarcity of cultured prostate cancer lines for testing. We previously developed an innovative method for generating immortal cultures from human prostatic epithelium which has proved uniformly successful in establishing over 20 new cell lines from benign and malignant prostatic tissue. Loss of allelic heterozygosity on chromosome 8p, the potential site of a suppresser gene related to prostate cancer, was used to characterize these lines. Using these lines as in vitro stimulants to raise tumor-reactive T cells from prostate cancer patients, we have identified prostate cancer-specific CD8+ T cells resticted in one case by HLA-B or -C, and in another case by a non-polymorphic MHC-like molecule. We are in the process of cloning the MHC class I-restricted prostate cancer Ag.The ultimate goal of these studies is to develop novel prostate cancer vaccines.